AI RESEARCH
Compact 3D Gaussian Splatting For Dense Visual SLAM
arXiv CS.CV
•
ArXi:2403.11247v3 Announce Type: replace Recent work has shown that 3D Gaussian-based SLAM enables high-quality reconstruction, accurate pose estimation, and real-time rendering of scenes. However, these approaches are built on a tremendous number of redundant 3D Gaussian ellipsoids, leading to high memory and storage costs, and slow